That is not good. That is not a load bearing outlet, that is for signal only. If you had the laser on a relay that might be okay.
I think that is rated at less than 0.2ma. I have to assume this is the issue, this would rob the ESP of power.
This is fine.
The laser is only on for about 10 seconds every 100 or so cuts. Its only on when I’m setting origin on irregular shapes. Once origin is set it turns off. Besides making a video I don’t think it’s been turned on in the last couple months. The spec sheet on the jackpot showed the 5v’s are rated for 25ma, the laser is running 25ma at max spec.
I jumped onto the jackpot page:
" * 2x 5V outputs
It’s being used as described.
At about 3:30 you can see what I’m doing.
So what tests might I run to cause an issue, any way to be sure it happens? I had it sitting here powered on for about 45 minutes yesterday and I could log in the whole time.
I don’t know. When I pulled it apart again it would respond to homing all, then trying to jog z a few times and it repeatedly failed.
It’s been running normally for quite a while, then it started not responding after cuts were completed until it was powered off and back on, then it was failing before cuts were made, then I couldn’t flash it, pretty much everything in this thread. I have no idea how to troubleshoot it, just that it wasn’t working correctly.
Well teh board terminals are kinda funky from use so I am not comfortable swapping boards, but if you are okay with it I will swap the ESP32 with the one I have been using on my LR, You have a generic ESP on there, since it was from so long ago I would be putting on a genuine.
That’s fine. It’s there any difference between this board and new? And can i un-funkify it?
Your jackpot screw terminals are chewed up from you using the wrong sized screwdriver.
Yes the one I will be putting on is a genuine esp32 not the knock off you paid for.
That’s only because my tiny flathead is not the super tiny one, only kind of tiny.
My wife says size doesn’t matter…
Ryan returned my board with a new ESP (thank you).
Unfortunately, if the ESP is good, the board is bad. Everything is put back together, connected and it made it through squaring the other day. Today I was leveling and it started into the exact same issues but with “error:152” in red. Multiple power on/off and it was then at least working for a bit so I decided to just go with it. I have cabinet doors off and my wife has hinted she would like them back on…
I was doing my 1st test cuts to measure square in actual use and it shut down entirely, displayed “websocket connection error” in red. The LED’s on the board were still lit up but all movement stopped. It had been actively moving for maybe 5 minutes.
It seems to connect initially, but after being on for a while starts having problems.
The only thing that I can think of, it’s 80-90 degrees in my shop.
Did Ryan have any batches with maybe a cold solder joint that might be opening up when it gets warm that could be re-flowed and hopefully fix this?
did you make or save any config edits? That is a configuration error and it was returned with a know proven good yaml. If you saved anything from the webui you can easily introduce an error.
AP or STA mode?
No, the only issue we ever had was a power socket overtightened and broke the through hole solder. Easy fix.
That was me 
Here’s the thing, I’ve been running this board since I bought it, and absolutely 0 was changed on the config once it was up and running square. So the errors were happening with no changes. I’ve had a lot of run-time on it and this is all new in the last month or so.
But since I got the board put back in:
I changed the pullback on Y1 to square the axis, doubled the steps on Z, my leadscrews are a different pitch, and added the pullback on the probe command so it works the same as the last interface, that is touch and retract a couple mm. That was all done through the configuration page on the web controller and saved by using their buttons. I haven’t tried to edit anything in a text editor.
The disconnects started during squaring. As before the wifi still shows it’s connected but it doesn’t respond to any controls. After a minute or 2 the web interface changes to the generic no network page, but wifi still shows it’s connected. I then connected by phone and got the same result.
Turning the machine/board off and back on, it reconnects and works normally for a few minutes, then repeated the disconnects. So it loads fine, then stops after a few minutes. I haven’t thrown a stopwatch on it to see if there’s any common run time before it happens, but that will likely happens soon. I don’t expect to get any usable info, but curiosity is what it is…
The disconnects continued while I was leveling the gantry, the only thing I changed was the z1 pullback after homing. During this process it was rehomed Z repeatedly, it was power cycled several times.
The websocket error popped up mid-cut, the 3rd test square, no power cycling so it was up and running for a few minutes. I’d done a 2x2 square to check diagonals, did a 2x8 rectangle and checked the diagonals, and was doing a 10x10 when it just shut down. There was a couple minutes between each cut while I measured and loaded the next file, Gcode generated by vectric vcarve pro. It had been up and running for several minutes, so the configuration had been loaded and had been untouched.
I’m in AP mode, connected directly to the board, not going through my home internet.
I was REALLY hoping to hear about a physical issue I could look at and hit with a soldering iron. I was thinking websocket was an actual socket on the board, google says I was thinking incorrectly.
I have it cutting air right now just to see if it will complete a file, I’ll check on it shortly.
Yep.
Reset again and I’m going to leave it powered up over night and see if it will stop just from on time or if it correlates to movement.
Did you ever confirm the details of how you’re powering the Jackpot?
It survived the night, it was still connected this morning. I jogged it twice, homed all, Z homed, Y homed X started homing and then it died, connection error, websocket error, connection with the board was lost.
24v power supply brick, same one I’ve been running since the initial primo build. I haven’t put a MM on it to check voltage but I’ve never seen the LEDs on the board flicker or go out so I assume it’s been receiving voltage the entire time.
editing
I’ll have to check its specs, I know it was rated for what was recommended, it was bought new just for the machine. I guess it might be failing under load, but the load on this thing is so low I’d be surprised. I don’t think I have another one laying around but I can dig through my ‘someday I might need’ pile of electrical parts and see if I do.
I would not be upset if a faulty power supply is the issue.
Slapping a multimeter on it would be a good idea. Making sure the wall plug to power supply box is good (we just had one that needed a pin bent out a bit), and looking at your through hole solder points is another check I already mentioned.
Having it fail for you during moves is a usually a smoking gun to something loose. Or something making contact that it should not.
Ryan tests these thoroughly, so if there is something wrong it is in the way the ESP-32 wifi stack, or FluidNC, or both interacts with your local network setup.
I changed out the power strip recently, but after it started having issues, it was part of my diy troubleshooting. Those plugs are tight and the strip was brand new.
The board itself has been in/out a few times now, so those connections have been moved repeatedly with no change in behavior. So power wise that leaves the ps itself, or the socket on the board.
Or am i missing one?
Ill hook up the MM later and see if i can get any variations when it’s running.
Just last week i came across a good deal on a dedicated PS and thought about picking it up but “I’m not building another machine for a while…”
I was on the verge of scrapping a somewhat expensive board, and it turned out to be the power supply. It would produce 25V when not under load, and could power the processor on board, but when it enaed the motors, thenPSU would sag badly, and everythjng would stop. I was pretty sure it was the motor drivers (onboard) so was about to replace it all.
Glad I checked with another PSU first.
